Virtual Machine Management Concurrency Limiting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualized computing environments, the allocation of hardware resources can lead to performance issues and inefficiencies due to overloading of networks and storage devices, resulting in slow response times and potential failures, especially when concurrency limits are exceeded, causing bottlenecks in virtual machine management operations.
Innovation Solution
Monitoring active virtual machine management operations across hosts to determine concurrency limits and initiating operations on hosts where these limits have not been met, allowing for the rescheduling of operations to avoid delays and errors by performing them on alternative hosts with available resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If virtual machine management operations are performed on a single host, then resource utilization is maximized, but concurrency limits are exceeded causing bottlenecks and delays
Solution Approach 1:
The patent segments virtual machine management operations into multiple concurrent sub-operations that can be distributed across different hosts. By dividing the workload and executing sub-operations in parallel on multiple hosts simultaneously, the system increases overall throughput while preventing any single host from exceeding its concurrency limits, thereby resolving the bottleneck problem.
Solution Approach 2:
The patent introduces a new dimension of host distribution for managing virtual machine operations. Instead of sequentially processing operations on a single host, the system distributes operations across multiple hosts in parallel, effectively adding a spatial dimension to operation execution. This dimensional change enables simultaneous processing without exceeding individual host concurrency limits.
2Productivity
If multiple sub-operations are executed concurrently on a host, then operational efficiency is improved, but concurrency limits are exceeded causing errors and failures
Solution Approach 1:
The patent implements a feedback mechanism that monitors the number of active sub-operations on each host and compares this against predefined concurrency limits. When the monitor detects that executing additional sub-operations would exceed the concurrency limit, it provides feedback to prevent further sub-operations from being scheduled on that host. This feedback loop ensures high utilization of parallel processing capability while maintaining operational reliability by preventing failures due to exceeded limits.
3Reliability
If virtual machine operations are delayed to avoid concurrency limits, then system stability is maintained, but throughput and performance deteriorate
Solution Approach 1:
The patent segments the workload across multiple hosts, allowing operations to proceed without delays by distributing them to available hosts. Instead of delaying operations when a host reaches its concurrency limit, the system divides and conquers by executing different sub-operations on different hosts simultaneously, maintaining both stability and high throughput.
Solution Approach 2:
The patent merges the capabilities of multiple hosts to handle virtual machine management operations. By combining the processing power and concurrency capacity of multiple hosts working in parallel, the system achieves higher overall throughput while maintaining stability, as each host operates within its own concurrency limits while contributing to the collective workload.
Data Source
AI summary
One or more concurrency limits may be checked in connection with the performance of a virtual machine management operation such as a virtual machine deploy, resize or migration operation to enable the virtual machine management operation to be scheduled on a host for which no concurrency limits have been met.